getting error in in tensorflow v 1.3

[Jumabek]

Please confirm, does this work later versions of TF as well?

I am getting following error when python run.py

feature dimensionality:  128
2017-12-07 17:34:08.821323: I C:\tf_jenkins\home\workspace\rel-win\M\windows-gpu\PY\36\tensorflow\core\common_runtime\gpu\gpu_device.cc:1045] Creating TensorFlow device (/gpu:0) -> (device: 0, name: GeForce GTX 1080, pci bus id: 0000:03:00.0)
2017-12-07 17:34:08.821453: I C:\tf_jenkins\home\workspace\rel-win\M\windows-gpu\PY\36\tensorflow\core\common_runtime\gpu\gpu_device.cc:1045] Creating TensorFlow device (/gpu:1) -> (device: 1, name: GeForce GTX 1080, pci bus id: 0000:04:00.0)
Press [ESC] to quit demo
2017-12-07 17:34:11.043341: E C:\tf_jenkins\home\workspace\rel-win\M\windows-gpu\PY\36\tensorflow\stream_executor\cuda\cuda_dnn.cc:371] could not create cudnn handle: CUDNN_STATUS_NOT_INITIALIZED
2017-12-07 17:34:11.043522: E C:\tf_jenkins\home\workspace\rel-win\M\windows-gpu\PY\36\tensorflow\stream_executor\cuda\cuda_dnn.cc:375] error retrieving driver version: Unimplemented: kernel reported driver version not implemented on Windows
2017-12-07 17:34:11.045137: E C:\tf_jenkins\home\workspace\rel-win\M\windows-gpu\PY\36\tensorflow\stream_executor\cuda\cuda_dnn.cc:338] could not destroy cudnn handle: CUDNN_STATUS_BAD_PARAM
2017-12-07 17:34:11.046425: F C:\tf_jenkins\home\workspace\rel-win\M\windows-gpu\PY\36\tensorflow\core\kernels\conv_ops.cc:672] Check failed: stream->parent()->GetConvolveAlgorithms( conv_parameters.ShouldIncludeWinogradNonfusedAlgo<T>(), &algorithms)

[obendidi]

the error that you get is not due to an incompatibility between the code and tensorflow version , but between your tensorflow installation and the version of cudnn that you installed

the code works with TF 1.3 and TF 1.4

[Jumabek]

ok Thanks !